Abstract. This paper describes a genetic algorithm (GA) tool added to
the MANA agent-based model to assist with scenario development. Squads
of agents are given chromosomes consisting of genes made up
from various personality weightings in the MANA model; the emphasis
is on evolving clever tactics and behaviour given the weapons
and equipment squads already have. Concepts from evolutionary biology such
as gene recombination and mutations are then applied to evolve
fittest squads to optimally defeat an enemy in a given
MANA scenario. We demonstrate the GA tool using two examples:
a simple shooting battle between two massed forces, and a
reconnaissance/counter-reconnaissance scenario in which a small Blue squad attempts to
locate a high value target within enemy territory. Communications links
in the MANA model are utilized for the information sharing,
thus highlighting issues of network enabled operations. Generally, the genetic
algorithm is seen to be a useful addition to the
toolkit of military modelling techniques based on complexity theory.
Related topics:
, training and analysis
View first page of "McIntosh: Genetic Algorithms Applied To Course-Of-Action Development Using The Mana Agent-Based Model"
Papers by McIntosh
Papers by Lauren
Register for the free
to receive a list of papers for each issue as it is released.
|
